Strategies to assess and manage hypervolemia The invisible threat in dialysis

Strategies to assess and manage hypervolemia The invisible threat in dialysis Rajiv Agarwal MD Professor of Medicine, Indiana University School of Medicine

Volume excess is common and costly. Admission for CHF in dialysis: 22% per year Volume overload needing acute dialysis: 14% per year Cost to Medicare (over 2.5 years): $266 million USRDS 2009 Annual Data Report Arneson TJ, et al. Clin J Am Soc Nephrol, 2010

HTN Volume Overload LVH CV Events Agarwal R, et al. Am J Med 115:291-297, 2003 Sarnak MJ, et al. Circulation 108:2154-2169, 2003 USRDS 2009 Annual Data Report Lazarus JM, et al. Arch Intern Med 133:1059-1066, 1974 Wizemann V, et al. Nephrol Dial Transplant 24:1574-1579, 2009 Agarwal R. Hypertension 56:512-517, 2010 MORTALITY

The challenge of assessing volume diagnosing hypervolemia

Seven methods to assess volemia History and examination Direct measurement of total body water Natriuretic peptides Bioimpedance analysis Relative plasma volume monitoring Blood pressure monitoring Echocardiogram

Pedal edema is not a volume marker Pedal edema is not associated with: IVC diameter BNP Relative plasma volume Inflammation markers Agarwal R, et al CJASN 3: 153, 2008

Odds ratio of Pedal Edema in HD Pedal edema associates with obesity, age, and LVH 400 350 300 250 BMI <25 kg/m2 (Normal or Underweight) BMI 25 to 29.9 kg/m2 (Overweight) BMI >30 kg/m2 (Obese) 200 150 100 50 0 LV mass < median (68.8 g/m2.7) LV mass median (68.8 g/m2.7) Agarwal R, et al CJASN 3: 153, 2008

Volume of distribution of alcohol = TBW

Drawbacks Several hours for measurement Alcohol exposure No data on how to use this information

In the DRIP trial, BNP did not Decline more with ultrafiltration Associate with more weight loss Associate with a greater improvement in interdialytic ambulatory BP

R ECW Adjusted for Bone Mineral Content 0 5 10 15 20 25 30 R ECW Adjusted for Lean Body Mass.4.6.8 1 1.2 1.4 Whole body BIA can detect changes in ECW Pre-HD Post-HD Pre-HD Post-HD Fisch BJ et al., Kidney Int 49: 1105, 1996

50 100 150 200 250 300 Resistance (ohms) Segmental resistance increases with UF-dialysis Pre-HD Post-HD Arm Trunk Leg Zhu F et al, ASAIO Journal 44: M541, 1998

Does BIA-guided dry-weight management result in better BP, TOD, outcomes? No study so far

Blood Volume Monitoring

Disposable Blood Chamber

Sensor Clip

Liquid Crystal Display

Full Assembly

RPV Principle Euvolemic Vascular Refill Increasing Hemo- Concentration Ultrafiltration Interstitial Space Vascular Space

80 85 90 95 100 105 Relative Plasma Volume (%) RPV Tracing 0 1 2 3 4 Time on Dialysis (hrs)

RPV Principle Hypervolemic Vascular Refill No Hemo- Concentration Ultrafiltration Interstitial Space Vascular Space

80 85 90 95 100 105 Relative Plasma Volume (%) RPV Tracing 0 1 2 3 4 Time on Dialysis (hrs)

RPV and Volemia Pre-specified secondary analysis of 150 HD patients in DRIP Trial

DRIP Trial Ambulatory BP RPV monitoring 150 patients 2 weeks On dialysis 3 months Ambulatory BP 135/85 8 weeks 100 patients ultrafiltration group Randomized 50 patients control group 8 weeks RPVM RPVM ABP ABP Agarwal R, et al. Hypertension 53:500-507, 2009

Study overview Design: Diagnostic test study Test: RPV monitoring Comparison standards of volume state: 1. Change in RPV slope 2. Ambulatory BP change

Relative Plasma Volume (%) Modeled RPV Slopes 100 98 96 Control Group Baseline Final Ultrafiltration Group Baseline Final 94 P < 0.001 92 0 1 2 3 4 Time on Dialysis (hrs) Sinha AD, et al. Hypertension 55:305-311, 2010

Change in Ambulatory BP (mmhg) +2 0-2 -4-6 Control Corrected Change in Ambulatory BP in Ultrafiltration Group +0.5-3.2-8 -10-12 P < 0.05-11.1-12.6-14 1 (steepest) 2 3 Quartile of RPV Slope 4 (flattest) Sinha AD, et al. Hypertension 55:305-311, 2010

Participants surviving (%) Ultrafiltration volume not predictive of mortality 25 50 75 100 0 6 12 18 24 30 36 42 48 54 60 66 Months since randomization Number at risk Higher UF Volume 151 133 120 99 82 70 52 42 39 36 23 14 Lower UF Volume 142 125 109 100 88 69 54 47 45 38 25 18 UF vol >2.7L UF vol <2.7L Logrank p = 0.208 Agarwal R, Hypertension 56: 512, 2010

Participants surviving (%) But flatter RPV slopes associated with a higher mortality 25 50 75 100 Number at risk Flatter slope Steeper slope Flatter Steeper Logrank p = 0.011 0 6 12 18 24 30 36 42 48 54 60 66 Months since randomization 155 136 121 103 87 73 52 44 39 34 22 15 154 138 123 111 97 80 66 55 51 43 29 18 Agarwal R, Hypertension 56: 512, 2010

Hazard Ratio for death Relative plasma volume is independently associated with mortality 3 2,5 2 1,5 Adjusted (age, sex, race, CVD, BP meds, dialysis vintage, albumin, Hgb) Adjusted + UF volume Adj + UFR Adj + UFR/kg Adj + UFR/kg + ABPM 1 0,5 0 p=0.02 p=0.008 p=0.002 p=0.001 P<0.001 Progessively Adjusted Models Agarwal R, Hypertension 56: 512, 2010

Does RPV-guided dry-weight management result in better BP, TOD, or outcomes?

RPV Monitoring Randomized Clinical Trial Crit-Line Intradialytic Monitoring Benefit (CLIMB) Study Multicenter trial with 6 centers in the U.S. and Canada Reddan DN, et al. J Am Soc Nephrol 16:2162-2169, 2005

CLIMB Study RPVM 6 months RPV monitoring 227 patients RPVM group 443 patients 2 weeks On dialysis 2 months Randomized 216 patients control group 6 months RPVM Reddan DN, et al. J Am Soc Nephrol 16:2162-2169, 2005

CLIMB Results RR for non-access related admit in RPVM group 1.49 (P = 0.017) Mortality 8.7% in RPVM group vs. 3.3% for control (P = 0.021) Reddan DN, et al. J Am Soc Nephrol 16:2162-2169, 2005

However Algorithm use was encouraged, but not mandated Highly variable implementation of the monitoring and interventional algorithm occurred within and across dialysis units. Reddan DN, et al. J Am Soc Nephrol 16:2162-2169, 2005

Control Group RPV Slope Baseline Distribution Steep 8% Flat 32% 60% Intermediate Reddan DN, et al. J Am Soc Nephrol 16:2162-2169, 2005

Control Group RPV Slope Final Distribution Steep 11% Flat 23% 65% Intermediate Reddan DN, et al. J Am Soc Nephrol 16:2162-2169, 2005

% of Patients with RPV slope Control group has fewer patients with flat slopes despite no intervention! 100% 80% 32 23 24 23 60% 40% 60 65 68 68 Flat Intermed Steep 20% 0% 8 11 8 9 Baseline 6 Months Baseline 6 Months Control Intervention Reddan DN, et al. J Am Soc Nephrol 16:2162-2169, 2005

Does RPV-guided dry-weight management result in better BP, TOD, or outcomes? Don t know, we need a better study!

Systolic Home BP of >150 mm Hg has a sensitivity of 80% and specificity of 84% in diagnosing interdialytic ambulatory hypertension 1.0 0.8 0.6 0.4 0.2 AUC 0.890 Home BP 0.0 0.0 0.2 0.4 0.6 0.8 1.0 Agarwal R, et al. Kidney Int 69: 900-906, 2006

Is out of dialysis unit BP of prognostic value? Baseline cohort followed for 2 years for all-cause mortality. 46 patients (31%) died. Quartiles of systolic BP associated with mortality in a Cox model. Alborzi P et al CJASN 2:1228, 2007

Hazard Ratio of All Cause Mortality Out of dialysis unit BP are of greater prognostic significance 3 2.5 Q1 Q2 Q3 Q4 Best home BP 125-145 mm Hg P=0.05 P=0.011 Best ABP 115-125 mm Hg 2 P=0.999 P=0.182 P=0.228 P=0.339 1.5 1 0.5 0 PreHD Routine Post HD Routine PreHD Standardized Post HD Standardized Home Ambulatory Alborzi P et al CJASN 2:1228, 2007

Does Home BP-guided BP management result in better BP, TOD, outcomes?

Home BP in HD Trial Design Home BP Group (n=34) Hypertension + Hemodialysis R Adjust antihypertensive medications Adjust antihypertensive medications 24-hr Ambulatory Blood Pressure + Echocardiogram Dialysis Unit Group (n=31) 0 6 months ABPM Echocard ABPM Echocard Da Silva GV, et al NDT 2009 doi: 10.1093/ndt/gfp332

24 hour ambulatory Systolic BP (mmhg) RCT shows that ambulatory BP improves with the use of home BP monitoring 175 165 Dialysis Unit Gp Home BP Gp 155 149 149 150 145 142 135 125 BL 6 mo No change seen on echo LVH Da Silva GV, et al NDT 2009 doi: 10.1093/ndt/gfp332

IVC Diameter End Expiration 1.89 cm = 10.5 mm/m 2

IVC Diameter End Inspiration 1.83 cm CI = 3.1%

Lung comets by ultrasound

Lung comets Jambrik Z et al. Am J Cardiol 93: 1265, 2004

Lung comets marker of pulmonary congestion correlated with ACM + cardiac events

What can you do today? Be aware of occult hypervolemia. Ask yourself the question: could these symptoms be due to hypervolemia? If discharged from the hospital, probe dry weight. If BP runs low and the patient is repeatedly hospitalized due to pneumonia, probe dry weight If home BP is high, probe dry weight Technology is unreliable, but RPV is the tool I use as my eyes to volume. Prescribe at least 4 hours of dialysis. If they cut their time or miss dialysis, they are likely not euvolemic. Root cause analysis: transportation, child-care, cramps etc Engage the dietitian, social worker, spouse, the patient to dialyze adequately.

Take home message Dialysis is not just about Kt/V, Hgb, Phos. Volume is important. We CAN do better!